On Performance Discrepancies Across Local Homophily Levels in Graph Neural Networks
On Performance Discrepancies Across Local Homophily Levels in Graph Neural Networks
复制标题
DOI：
10.48550/arxiv.2306.05557
复制
发表时间：
2023-06
期刊：
影响因子：
--
通讯作者：
Donald Loveland;Jiong Zhu;Mark Heimann;Benjamin Fish;Michael T. Shaub;Danai Koutra
中科院分区：
文献类型：
--
作者：
Donald Loveland;Jiong Zhu;Mark Heimann;Benjamin Fish;Michael T. Shaub;Danai Koutra
图神经网络（GNN）研究强调了节点分类中高同质性（即同一类节点连接的趋势）与强预测性能之间的关系。然而，最近的研究发现这种关系更加微妙，表明简单的 GNN 可以在某些异质环境中学习。为了解决这些相互矛盾的发现并更接近现实世界的数据集，我们超越了全局图同质水平的假设，并研究了当节点的局部同质水平偏离全局同质水平时 GNN 的性能。通过理论和实证分析，我们系统地证明了局部同质性的变化如何导致性能下降，从而导致局部同质性水平之间的性能差异。我们通过对具有不同全局同质性水平的五个现实世界数据集进行粒度分析，证明了这项工作的实际意义：(a) GNN 可能无法泛化到偏离图的全局同质性的测试节点，(b) 高局部同质性并不一定能为节点带来高性能。我们进一步表明，为全局异质图设计的 GNN 可以通过提高局部同质水平的性能来减轻性能差异，为这些 GNN 如何实现更强的全局性能提供了新的视角。
Graph Neural Network (GNN) research has highlighted a relationship between high homophily (i.e., the tendency of nodes of the same class to connect) and strong predictive performance in node classification. However, recent work has found the relationship to be more nuanced, demonstrating that simple GNNs can learn in certain heterophilous settings. To resolve these conflicting findings and align closer to real-world datasets, we go beyond the assumption of a global graph homophily level and study the performance of GNNs when the local homophily level of a node deviates from the global homophily level. Through theoretical and empirical analysis, we systematically demonstrate how shifts in local homophily can introduce performance degradation, leading to performance discrepancies across local homophily levels. We ground the practical implications of this work through granular analysis on five real-world datasets with varying global homophily levels, demonstrating that (a) GNNs can fail to generalize to test nodes that deviate from the global homophily of a graph, and (b) high local homophily does not necessarily confer high performance for a node. We further show that GNNs designed for globally heterophilous graphs can alleviate performance discrepancy by improving performance across local homophily levels, offering a new perspective on how these GNNs achieve stronger global performance.